Confocal Scanner System for Long-term Live Cell Imaging Confocal

Confocal Scanner System for Long-term Live Cell Imaging

Yokogawa Technical Report English Edition Vol.55 No.1 (2012)

Confocal Scanner System for Long-term Live Cell ImagingHironori Takai *1 Kenji Hachiya *1

Youhei Kuwabara *2 Takayuki Kei *1

The CV1000 is a new confocal microscopy system that enables multi-point, long-term 3D imaging of live cells while keeping them healthy and active. Based on Yokogawa’s CSU confocal scanner, the all-in-one CV1000 is equipped with laser light sources, highly sensitive EMCCD camera, precision XYZ auto stage, auto-focus and incubator functions. This system can be widely used by many researchers, especially for long-term observation of the development process of mammalian embryos, colony formation process of iPS/ES cells, and so on.

INTRODUCTION

As f luorescent proteins including green f luorescent protein (GFP) have been put into practice, biotechnology

research that employs imaging that uses them has progressed greatly. In particular, observation with a confocal system has become popular for its clear images, yielding many successful results (1), (2). Yokogawa’s CSU series confocal scanners (3) have been widely used as fluorescent protein gains popularity and are now well-recognized as a useful tool for observing live cells.

The major reason why the CSU ser ies has gained popularity in the market is its high-speed processing capability and low phototoxicity to cells. Combining these optical technologies in the CSU series with Yokogawa’s excellent environmental control and precise positioning technologies, the Life Science Headquarters has developed the CellVoyager CV1000 confocal scanner system to observe live cells. This product can be used not only for biological basic research but also for research in new drug development and regenerative medicine. We aim to make this product a de facto standard

tool in biotechnology.The CV1000 satisfies all the following requirements for

observing live cells. • Extremely low phototoxicity to cells • High-speed processing to track the behavior of live cells • Accuracy and high reproducibility • Robustness and reliability for long-term usage • Easy-to-use softwareThe main specifications of the unit are as follows: • Excitation light source: Three-color laser (405 nm, 488

nm, and 561 nm), LED transmission for the bright field • Confocal unit: Pinhole disk with microlens (φ25 µm and φ50 µm)

• Objective lens: Max. 6 lenses (4×, 10×, 20×, 40×, 60×, and 100×)

• Auto-focus: By laser • Fluorescent filter wheel: 6 positions • Camera: Back-illuminated EMCCD camera • Temperature control unit: 30°C to 40°C (normal

temperature), 17°C to 30°C (low temperature) • CO2 concentration control unit: 5% • Humidity control: Forced humidification • Control softwareT h is pape r out l i nes t he CV10 0 0 and desc r ibes

characteristics of its hardware and software.

27 27

*1 Development Dept., Life Science Headquarters*2 Sales Dept., Life Science Headquarters


Yokogawa Technical Report English Edition Vol.55 No.1 (2012) 2828

OUTLINE AND CONFIGURATION OF CV1000

Figure 1 shows the external view of the CV1000, consisting of the main body, power supply unit, and PC for control. The main body consists of three parts: the environmental control system for culturing cells; the optical system for a confocal microscope; and the mechatronic part including the XY stage for conveying specimens and the Z stage for adjusting the objective lens position. The CV1000 supports various culturing containers such as slide glasses, cover glass chambers, glass bottom dishes, and micro plates.

Figure 1 External view of the CV1000

The CV1000 is a compact all-in-one system. The observation part is completely shielded from outside light, requiring no dark room facility. Media for cell cultures can be exchanged without opening the cover. The CV1000 can keep a stable environment to perform experiments for up to one week.

Conventional confocal microscopes require much effort and know-how for everything from installation to operation, which depends on a user’s skills. The CV1000 is a turn-key system which requires no operational expertise in hardware and software for operation, and can be used in various researches.

SoftwareFigure 2 shows a screen of the CV1000 control software

for capturing images and controlling equipment. This screen is designed so that users do not need to pay attention to hardware. Users can quickly obtain images by clicking a single button even with a default configuration.

The control software has various features for setting experimental conditions and imaging procedures, editing obtained images, creating movies and the like. Even during imaging, users can easily select a specimen from a preview map, obtain the whole view of a specimen in culturing containers such as dishes, select and observe multipoint images, and check obtained images.

Thanks to the high-speed, high-precision XYZ stage, imaging can be performed quite smoothly while changing observation views and selecting the position in depth direction

.

Figure 2 A screen of the CV1000 control software

The platform is Microsoft Windows 7 64-bit version with .NET 4.0 Framework. On a 64-bit operating system, the software can use more than 2 GB memory to easily process a 2 MB three-dimensional stack image output from a 1-megapixel 16-bit color camera.

Technologies such as the C# 3.0, Windows Presentation Foundation (WPF), and Shader Effect are used. Language Integrated Query (LINQ) in C# 3.0 helps the software to be written in declare statements, making it less complex. WPF is used as a GUI platform to enhance GUI expression and f lexibility. The Shader Effect is used for image processing calculation. It enables high-speed calculation with a graphic card to process images obtained at four different wavelengths and display their merged image.

For the software architecture, we use the Model View ViewModel (MVVM) pattern as shown in Figure 3. By using data binding functions in WPF, we have developed the software with sophisticated GUI while keeping loose the coupling between GUI and logic.

This architecture helps ensure easy maintenance of software and improve the design quality.

Figure 3 The outline of the MVVM pattern

The optical system for the confocal microscopeFigure 4 shows the optical system of the CV1000. As

an excitation light source, lasers with three wavelengths (405 nm, 488 nm, and 561 nm) are provided to support popular f luorescent dye and f luorescent proteins such as Hoechst, DAPI, GFP, and mREP. These three laser beams are multiplexed, paralleled through the collimate lens, and enter the confocal scanner. Through the microlens array, these parallel lights become multiple point lights, and then each point light passes through the corresponding pinhole. Then, the point lights pass through the tube lens and objective lens and focus on the specimen to excite cells. When fluorescence is excited from a fluorescent-labeled target in the specimen, the fluorescence passes through the objective lens and tube lens in the opposite direction of the excitation light and enters the confocal scanner. After passing through the pinhole array,

Main body

Power unit

View ModelViewModel

Presentation Business Logicand Data

DataBinding

Command

User Interface



the f luorescence signals form a confocal image, which is reflected by the dichroic mirror (DM), passes through the relay lens, and forms an image on the camera.

Dif ferent f luorescent-labeled proteins a re used for respective observation targets. Exciting lights with wavelengths corresponding to respective f luorescents are selectively turned on and off, and at the same time, the fluorescent filter in front of the camera is switched, and then pictures are sequentially taken for each wavelength. Finally, those single-color images are superimposed to create a multicolor image.

To observe the three-dimensional structure of cells, the Z stage is used to move the objective lens along the light axis, and cross-sectional images of cells are obtained for each submicron-to-few-micron movement. Then, those images are stacked to create a three-dimensional image.

To observe the appearance of cells, LED transmission is used for a bright field light source.

To support various observations from a wide view to small targets such as endoplasmic reticulum in cells, this system has six objective lenses ranging from 4× to 100×, which can be electrically switched.

Figure 4 The optical system of the CV1000

The confocal scanner with selectable pinhole sizeThe major advantage of confocal systems is higher

resolution in the Z-direction (the direction of depth of the specimen) than that of epi-fluorescence systems. The higher resolution in the Z-direction helps obtain thinner optical slices and thus the fine structure of cells can be observed. The thickness of the optical slice is determined by the following equation (1) with variables such as the pinhole diameter, numerical aperture (NA) of objective lens, and wavelength of fluorescence(4). This equation clearly shows that a thinner optical slice can be obtained by an objective lens with a larger NA and pinholes with a smaller diameter.

Where,lem: Fluorescence wavelength

n: Refractive index of medium between the objective lens and the specimen container

NA: Numerical aperture of the objective lensPH: Pinhole diameterFigure 5 shows the thickness of an optical slice obtained

by the equation (1) by substituting the wavelength of GFP (520 nm) for lem and numerical apertures of typical objective lens for NA. The lower the magnification the objective lens has and the smaller the diameter the pinhole has, the thinner an optical slice can be obtained. In other words, larger confocal effects are very effective.

Figure 5 The relationship between pinhole diameter and optical slice thickness (l: 520 nm)

Based on the results shown in Figure 5, Yokogawa’s CSU confocal scanners use objective lenses with high magnification alone. The diameter of pinholes is set to 50 µm; due to a limitation to the structure of scanners, it is difficult to change this value. When a lens with low magnification is used for morphological observation, the confocal system does not exert its full effects. Therefore, confocal scanners are mainly used for research that needs high magnification. To overcome the limitation, the CV1000 has two disks with pinholes of 50 µm and 25 µm respectively to satisfy a strong demand for diameter selectivity. Overcoming the disadvantage of the conventional CSU, a single CV1000 exerts full confocal effect with various objective lenses from low to high magnification.

Figure 6 shows the actual measurement result of the point spread function for different pinhole diameters when the objective lens (10×, NA = 0.4) is used. The point image of the 25 µm pinhole is about 15% narrower than that of 50 µm, showing the superiority of small pinholes. In addition, smaller pinholes reduce crosstalk caused by lights from other than the focal plane and improve the S/N ratio of confocal images.

Figure 6 The point spread function (PSF) of the CV1000

405 nm

Pinhole array disk(with φ50 µm pinholes)

Beam combiner

Transmission light source

Fluorescent filter

Specimen

Dichroic mirror (DM)

Pinhole array disk(with φ25 µm pinholes)

Objective lens

XY stage

Relay lens

Imaging lens

Confocalscanner

488 nm

561 nm

Camera

Relay lensZ stage

Collimatelens

Chamber

Dark-room structure

Microlens array disk

· · · · · · (1)22

22

288.0

⋅⋅+

−−=

NAPHn

NAnnd emλ

0

2

4

6

8

10

0 10 20 30 40 50 60 70 80

100×Oil/1.460×W/1.240×/0.9520×/0.7510×/0.4

Pinhole diameter [µm]

Opt

ical

slic

eth

ickn

ess

[µm

]

0.0

0.5

1.0

-20 -15 -10 -5 0 5 10 15 20

25µm Pinhole50µm Pinhole

Optical axis direction [µm]

Rel

ativ

e In

tens

ity

29 29



Figure 7 shows sample images of cells. The Z-section image of φ25 µm pinholes is sharper than that of φ50 µm pinholes.

Figure 7 Sample images of cells (10×, NA = 0.4)

High-precision environmental controlIn the developmental process research, samples such as

fertilized eggs and early embryos are incubated and observed over a dozen days. Since these samples are sensitive to the ambient environment, the success of the experiment depends heavily on the environmental control.

For the CV1000, a highly stable environmental chamber has been developed to keep the temperatures between 30°C and 40°C, humidity at 90% or more, and CO2 concentration at 5% in the measurement part. For example, a chamber for a single dish can keep the temperature in the above-mentioned range with a fluctuation of ±0.1°C. To ensure this performance, we have made various improvements. Specif ically, the housing is covered with a thick heat insulator to reduce the impact from the outside temperature fluctuation, making the inside environment similar to that in a thermostatic chamber. Furthermore, to curb temperature fluctuation in the chamber, the chamber is divided into several zones from top to bottom and respective temperatures are independently controlled by heaters and temperature sensors mounted in each zone. The temperature of the objective lenses is also controlled individually to eliminate even a slight temperature change during replacement.

Figure 8 shows an example of temperature control. After the transition period, the fluctuation of the temperature in the chamber is controlled to within ±0.1°C, which ensures the high temperature stability in the observation part.

Figure 9 shows an example of CO2 concentration control in the chamber. When culturing cells, the ambient CO2 concentration is usually controlled at 5% to keep the pH value

of the medium constant. 100% CO2 is mixed with air and the concentration is controlled by using a near-infrared gas sensor. The concentration becomes stable about five minutes after the start of control.

Figure 8 Example of temperature control

Figure 9 Example of the CO2 concentration control

CONCLUSION

Bio-imaging technologies for visualizing minute behavior in cells have contributed a lot to clarifying life phenomena. The need for higher-speed, more precise, and longer-term observation of behavior of proteins in live cells is sure to increase. Yokogawa will respond to this need and offer ideal devices for research.

REFERENCES

(1) K. Matsuura-Tokita, M. Takeuchi, et al., “Live imaging of yeast Golgi cisternal maturation,” Nature, Vol. 441, No. 7096, 2006, pp. 1007-1010

(2) K. Yamagata, R. Suetsugu, et al., “Long-term, six-dimensional live-cell imaging for the mouse preimplantation embryo that does not affect full-term development,” J Reprod Dev, Vol. 55, No. 3, 2009, pp. 343-350

(3) Hideo Hirukawa, Hiroshi Nakayama, et al., “New Technologies for CSU-X1 Confocal Scanner Unit,” Yokogawa Technical Report English Edition, No. 45, 2008, pp. 21-26

(4) S. Wilhelm, B. Grobler, M. Gluch, H. Heinz, “Confocal Laser Scanning Microscopy : Optical Image Formation and Electronic Signal Processing,” Carl Zeiss Jena GmbH, 45-0029 e/09.03, pp. 15

* CSU is a registered trademark of Yokogawa Electric Corporation.* Cell Voyager is a registered t rademark of Yokogawa Elect r ic

Corporation.* CV1000 is a registered trademark of Yokogawa Electric Corporation.

(b) Z-section at the arrowed line with ϕ50 µm pinholes

(a) XY-plain image of mouse jejunum sample

(b) Z-section at the arrowed line with ϕ25 µm pinholes

36.2

36.6

37.0

37.4

37.8

0 15 30 45

Tem

pera

ture

[°C

]

Time [min]

0

5

10

0 5 10 15 20Time [min]

CO

2 Con

cent

ratio

n [%

]

3030

Documents

Confocal Scanner System for Long-term Live Cell Imaging Confocal